From d46165cacba88867517b1661474d0265e10693eb Mon Sep 17 00:00:00 2001
From: James Healy <jimmy@deefa.com>
Date: Wed, 6 Jul 2011 19:51:20 +1000
Subject: [PATCH] support multiple receivers in a single parse of a source PDF

* In cases were multiple receivers are needed, this saves a *huge*
  amount of time
* before this, the source file was completely parsed and extracted N
  times for N receivers
* now it's parsed and extracted 1 time for N receivers
---
 CHANGELOG                           |  4 ++++
 lib/pdf/reader.rb                   | 23 +++++++++++++++++------
 lib/pdf/reader/abstract_strategy.rb | 18 ++++++++++--------
 spec/pages_strategy_spec.rb         | 18 ++++++++++++++++++
 4 files changed, 49 insertions(+), 14 deletions(-)

diff --git a/CHANGELOG b/CHANGELOG
index c81e6611..6a2e120c 100644
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,7 @@
+v0.9.4 (XXX)
+- support multiple receivers within a single pass over a source file
+  - massive time saving when dealing with multiple receivers
+
 v0.9.3 (2nd July 2011)
 - add PDF::Reader::Reference#hash method
   - improves behaviour of Reference objects when tehy're used as Hash keys
diff --git a/lib/pdf/reader.rb b/lib/pdf/reader.rb
index 0278983f..fdea1130 100644
--- a/lib/pdf/reader.rb
+++ b/lib/pdf/reader.rb
@@ -73,21 +73,32 @@ module PDF
   #   :pages
   #   :raw_text
   #
+  # = Processing with multiple receivers
+  #
+  # If you wish to parse a PDF file with multiple simultaneous receivers, just
+  # pass an array of receivers as the second argument:
+  #
+  #   pdf = PDF::Reader.new
+  #   pdf.parse(File.new("somefile.pdf"), [receiver_one, receiever_two])
+  #
+  # This saves a significant amount of time by limiting the work to a single pass
+  # over the source file.
+  #
   class Reader
 
     # Parse the file with the given name, sending events to the given receiver.
     #
-    def self.file(name, receiver, opts = {})
+    def self.file(name, receivers, opts = {})
       File.open(name,"rb") do |f|
-        new.parse(f, receiver, opts)
+        new.parse(f, receivers, opts)
       end
     end
 
     # Parse the given string, sending events to the given receiver.
     #
-    def self.string(str, receiver, opts = {})
+    def self.string(str, receivers, opts = {})
       StringIO.open(str) do |s|
-        new.parse(s, receiver, opts)
+        new.parse(s, receivers, opts)
       end
     end
 
@@ -111,7 +122,7 @@ def self.object_string(str, id, gen = 0)
 
     # Given an IO object that contains PDF data, parse it.
     #
-    def parse(io, receiver, opts = {})
+    def parse(io, receivers, opts = {})
       ohash    = ObjectHash.new(io)
 
       if ohash.trailer[:Encrypt]
@@ -122,7 +133,7 @@ def parse(io, receiver, opts = {})
       options.merge!(opts)
 
       strategies.each do |s|
-        s.new(ohash, receiver, options).process
+        s.new(ohash, receivers, options).process
       end
 
       self
diff --git a/lib/pdf/reader/abstract_strategy.rb b/lib/pdf/reader/abstract_strategy.rb
index 83bff559..06826eb8 100644
--- a/lib/pdf/reader/abstract_strategy.rb
+++ b/lib/pdf/reader/abstract_strategy.rb
@@ -4,8 +4,13 @@ class PDF::Reader
 
   class AbstractStrategy # :nodoc:
 
-    def initialize(ohash, receiver, options = {})
-      @ohash, @receiver, @options = ohash, receiver, options
+    def initialize(ohash, receivers, options = {})
+      @ohash, @options = ohash, options
+      if receivers.is_a?(Array)
+        @receivers = receivers
+      else
+        @receivers = [receivers]
+      end
     end
 
     private
@@ -17,7 +22,9 @@ def options
     # calls the name callback method on the receiver class with params as the arguments
     #
     def callback (name, params=[])
-      receiver.send(name, *params) if receiver.respond_to?(name)
+      @receivers.each do |receiver|
+        receiver.send(name, *params) if receiver.respond_to?(name)
+      end
     end
 
     # strings outside of page content should be in either PDFDocEncoding or UTF-16.
@@ -56,10 +63,6 @@ def pages?
       pages ? true : false
     end
 
-    def receiver
-      @receiver
-    end
-
     def root
       ohash.object(trailer[:Root])
     end
@@ -74,4 +77,3 @@ def trailer
 
   end
 end
-
diff --git a/spec/pages_strategy_spec.rb b/spec/pages_strategy_spec.rb
index 5d4fb2a8..74cfba16 100644
--- a/spec/pages_strategy_spec.rb
+++ b/spec/pages_strategy_spec.rb
@@ -87,4 +87,22 @@ class PDF::Reader::PagesStrategy
     text_callbacks[0][:args].should eql([["My name is"]])
     text_callbacks[1][:args].should eql([["James Healy"]])
   end
+
+  it "should send the correct callbacks when using more than one receiver" do
+
+    # mock up an object that will be called with callbacks. This will test that
+    # the content class correctly recognises all instructions
+    one = mock("receiver_one")
+    one.should_receive(:move_text_position).once # Td
+
+    two = mock("receiver_two")
+    two.should_receive(:move_text_position).once # Td
+
+    # The instructions to test with
+    instructions = "36.000 794.330 Td"
+
+    # process the instructions
+    content = PDF::Reader::PagesStrategy.new(nil, [one, two])
+    content.content_stream(instructions)
+  end
 end